Nitrogen 1s binding energies of transition metal nitrosyls - Inorganic

Patricia. Finn, and William L. Jolly. Inorg. Chem. , 1972, 11 (4), pp 893–895. DOI: 10.1021/ic50110a051 .... William B. Hughes and Bernard A. Baldwi...
0 downloads 0 Views 457KB Size
Inorganic Chemistry, VoZ. 11, No. 4 , 1972 893

NOTES to the diethylamino derivative upon reaction with diethylamine. In the reactions of the chlorophosphole with AgNCO, AgNCS, C6H5SH, CH3SH, and CH30HJ4spectroscopic data indicated that the isolated products were, respectively, I, where X = NCO, NCS, SC6H5, SC&, and OCH3. However, these phospholes changed their identity rapidly, and we were unable to obtain pure samples for satisfactory identification and further reaction with W(CO)5.THF. The literature contains many examples of the lability of phospholes toward Arbuzov and similar reactions5 Attempts to stabilize the labile ligands, by immediate reaction with the tetrahydrofuran complex of tungsten pentacarbonyl, in many cases gave isolatable complexes which decomposed rapidly both in vacuo and under nitrogen to give uncharacterized blue residues. In contrast, the new compounds CsH402P(C1)W(CO)5 and C6H402P [N(CzH5)2]W(C0)5have been stable for up to 2 years in a nitrogen atmosphere. The reactions of the chlorophosphole with silver cyanide (in dichloromethane and in xylene, under reflux) and with methyl- or phenylmagnesium bromide did not yield the desired derivatives of I. Furthermore, we were unable t o obtain a pure sample of the phenylphosphole from the reaction of phenylphosphorus dichloride with catech01.~~' The proton nmr spectrum of the aminophospholetungsten pentacarbonyl complex, when compared t o that for the uncomplexed ligand, shows a downfield shift for all the ligand protons consistent with a formal oxidation state change from 3+ to 5+ for phosphorus upon complexation (assuming that phosphorus is coordinating to tungsten), Experimental Section Physical Measurements.-Infrared spectra were determined using a Perkin-Elmer Model 21 sodium chloride spectrophotometer and a Perkin-Elmer 621 grating spectrometer. Proton nrnr spectra were obtained by means of a Varian A-60 spectrometer using tetramethylsilane as an internal standard. Mass spectra were taken using a Perkin-Elmer Hitachi RMU-BD doublefocusing spectrometer a t 70-eV ionizing energy. Irradiation was carried out in a quartz vessel using a Hanovia 679A36 medium-pressure lamp. Analyses were carried out by Schwarzkopf Microanalytical Laboratories, Woodside, N Y ., and Micro-Tech Laboratories, Inc., Skokie, Ill. Materials.-Tungsten hexacarbonyl was a generous gift from the Climax Molybdenum Co. 2-Chloro-1,3,2-benzodioxaphosphole was prepared by a literature method4 and all solvents were dried before use by standard procedures. All manipulations were carried out under nitrogen. Preparation of CeH40zP(C1)W(CO)a(II).-A solution of 1.76 g (5 mmol) of tungsten hexacarbonyl in 50 ml of tetrahydrofuran was irradiated for 5 hr, during which time 5 mmol of carbon monoxide was evolved. The resulting yellow solution was added quickly to 0.87 g (5 mmol) of C ~ H ~ O Z and P C ~the mixture was stirred for 20 min a t ambient temperature. Solvent was removed in vacuo and the residual yellow solid sublimed. Unreacted tungsten hexacarbonyl (0.35 g) was collected a t 30" (0.05 mm) and identified by it sinfrared spectrum. Fine colorless needles of product (0.75 g, 407,) were collected at 50-55' (0.05 rnm), mp 80-81O. A n a l . Calcd for CI1H4C1O,PW: C, 26.5; H, 0.8; C1, 7.1. Found: C, 26.5; H, 1.1; C1, 7.2. Ir spectrum (hexane solution): v ( C 0 ) 2083 (m), 1981 (vs), 1950 (m sh) cm-l. A parent ion was observed in the mass spectrum at 500 (calcd: 500 for ls4W and 37Cl).

.

(4) A. E. Arbuzov and F. G. Valitova, Bull. Acad. USSR, Div. Chem. Sci., 529 (1940);Chem. Abstr., 86, 3690 (1941). (5) R. G. Harvey bud E. R . DeSombre, Top. Phosphoriis Chem., 1, 67 (1964). (6) K.D.Berlin and M. Nagabhushanam, J . Org. Chem., 29, 2056 (1964). (7) M.Wieber and W, R . Hoos, Tefvahedron Left., 61, 5333 (1968).

Preparation of 2-Diethylamino-~,3,2-benzodioxaphosphole, c6H~OZPN(CZH~)Z.-A 2.9-g (40-mmol) sample of diethylamine in 20 ml of benzene was added dropwise, over a 4-hr period, to 3.5 g (20 rqmol) of 2-chloro-l,3,2-benzodioxaphospholein 20 ml of benzene, cooled to 0" in a two-necked flask equipped with a mechanical stirrer. Vigorous stirring was maintained for a further 16 hr a t ambient temperature under nitrogen. Solvent was removed zn vacuo and the residue was shaken with pentane. The mjxture was filtered, solvent was removed, and the residual oil was distilled under reduced pressure. The product (1.9 g, 46%) was collected a t 113-115' (0.25 mm). Anal. Calcd for CloH14NOzP: C, 56.9; H , 6.6; N, 6.6; P , 14.7. Found: C, 58.4; H , 6.9; N, 6.4; P, 14.0. Proton nmr (carbon tetrachloride solution): T 3.23 (four phenyl protons), 7.14 (six methylene protons), 9.02 (six methyl protons). Preparatibn of C S H ~ O ~ P [ N ( C ZW(C0 H S ) ~ ])j.-The preparation was essentially the same as for I1 above [0.88 g (2.5 mmol) of tungsten hexacarbonyl, 7 hr irradiation, and 0.52 g (2.5 rnmol) of ligand]. After removal of solvent the pink solid was sublimed t o give 0.13 g of unreacted tungsten hexacarbonyl (40' (0.05mm)) and fine colorless needles of product (0.76 g, 677,) a t 75" (0.05 mm), mp 67'. Anal. Calcd for C I ~ H I ~ N O ~ P W C, : 33.7; H , 2.6; P , 5.8; W, 34.4. Found: C, 33.9; H , 2.6; P , 5.9; W, 33.3 Ir spectrum (hexane solution): v ( C 0 ) 2081 (m), 1962 (vs) cm-'. A parent ion was observed in the mass spectrum a t 535 (calcd: 535 for lsrW). Proton nmr (carbon tetrachloride solution): 7 3.01 (four phenyl protons), 6.82 (four methylene protons), 8.84 (six methyl protons). Reaction of 2-Chloro-1,3,2-benzodioxaphospholewith Silver 1sothiocyanate.-A 8.73-g (50-mmol) sample of 2-chloro-l,3,2benzodioxaphosphole and 8.30 g (50 mmol) of silver isothiocyanate in 100 ml of benzene were heated under reflux for 16 hr , The cooled mixture was filtered, solvent was removed in vacuo, and the residual viscous liquid was distilled. The main fraction (5.15 g) was collected a t 70' (1.5 mm). The mass spectrum showed a parent ion a t 197 which agrees with the calculated value of 197 for C ~ H ~ N O Z P SHowever, . this product rapidly turned into a yellow-orange oil upon standing a t room temperature, in vacuo, in the dark.

CONTRIBUTION FROM THE DEPARTMENT OF CHEMISTRY OF THE UNIVERSITY OF CALIFORNIA AND THE INORGANIC MATERIALS RESEARCH DIVISIONOF THE LAWRENCE BERKELEY LABORATORY, BERKELEY, CALIFORNIA94720

The Nitrogen Is Binding Energies of Transition Metal Nitrosyls BY PATRICIA FINNAND WILLIAML. JOLLY*

Received July 15, 1971

The bonding and electronic structure of transition metal nitrosyls have been frequently studied in recent years. 1-6 Two limiting situations have been identified: linearly coordinated NO+ groups and angularly coordinated ("bent") NO- groups. We have investigated the X-ray photoelectron spectra7 of a series of these compounds to attempt correlations of the nitro(1) B. F. G. Johnson and J. A. McCleverty, Pvogv. Inoug. Chem., 7, 277 (1966). (2) P.Gans, Chem. Commun., 144 (1968). (3) D. J. Hodgson, N. C. Payne, J. A. McGinnety, R. G. Pearson, and J. A. Ibers, J . Amev. Chem. SOL,90, 4486 (1968). (4) D.J. Hodgson and J. A. Ibers, Inoug. Chem., 7, 2345 (1968). (5) D.J. Hodgson and J. A. Ibers, ibid., 8, 1282 (1969). (6) D.M. P. Mingos and 5. A. Ibers, ibid.,10, 1479 (1971). (7) K. Siegbahn, C. Nordling, A. Fahlman, R. Nordberg, K . Hamrin, J. Hedman, G. Johannson, T. Bergmark, S. E. Karlsson, I. Lindgren, and B. Lindberg, "ESCA Atomic, Molecular and Solid State Structure Studied by Means of Electron Spectroscopy," Almqvist and Wiksells AB, Stock-

holm, 1967.

894 Inorganic Chemistry, Vol. 11,No. 4, 1972 gen 1s electron binding energy with s t r u c t ~ r e , elec~-~~ tronic features, and N-0 stretching frequency. 16,20--39 Experimental Section The compounds were kindly supplied by Philip G. Douglas and Robert D. Feltham of the University of Arizona, by G . Dolcetti, P . Farnham, and James P. Collman of Stanford University, and, in the case of [Co(NHa)sNO]CIZ,~O by Mark Iannone of this department. The powdered samples were brushed onto double-faced conducting tape attached to a n aluminum plate. The reproducibility was 2 ~ 0 . 2eV. I n each case the carbon Is line (due to a film of pump oil on the samples) was recorded and used as a reference peak. Individual lines had widths a t half-height of 1.2-2.8 eV; broader lines were due to decomposition or to the presence of a second peak. T h e kinetic energy of the photoelectron, E K ,was measured in a n iron-free double-focusing magnetic spectrometer41 in which the incident radiation, Ehv, was magnesium Ka: X-radiation (1253.6 eV) and for which the work function, &, was taken as 4.0 eV. The nitrogen 1s binding energy, E B (the difference between the Fermi level and the 1s atomic level energy), was calculated from the relation E B = Ehv - E K - &.

Results and Discussion The data are presented in Table I. I t was found advantageous to categorize the compounds in terms of the metal d-electron configurations by making the arbitrary assumption that the nitrosyl groups were NO+ ions. (Although this assumption is illogical for the few compounds that contain “bent” NO groups, we made the assumption to simplify the classification of the compounds.) The first seven compounds in Table I are six-co(8) J. H. Enemark, M .S. Quinby, L. L. Reed, M. J. Steuck, and K . K. Walthers, Inorg. Chem., 9, 2397 (1970). (9) A. Tulberg and N. G. Vannerberg, Acta Chem. Sca?zd., 21, 1462 (1967). (10) P . T. Manoharan and W. C. Hamilton, I n o u g . Chem., 2, 1043 (1963). (11) D. M. P. Mingos, W.T. Robinson, and J. A. Ihers, zbid., 10, 1043 (1971). (12) D. M. P. Mingos and J. A. Ibers, i b i d . , 10, 1035 (1971). (13) C. S. Pratt, B. A. Coyle, and J. A. Ibers, J . Chem. Soc., 2146 (1971). (14) D. A. Snyder and D . L. Weaver, I n o r g . Chem., 9, 2760 11970). , (15) C. G. Pierpont, A. Pucci, and R . Eisenberg, J . Amer. Chem. S O L .93, 3050 (1971). (16) C. G. Pierpont, D. G. Van Derveer, W. Durland, and R. Eisenberg, i b i d . , 92, 4760 (1970). (17) D . M . P. Mingos and J. A. Ibers, Illorg. Chem., 9, 1105 (1970). (18) V, G. Albano, P. L. Bellon, and M. Samoni, quoted in Table VI1 of ref 6. (19) J. H. Enemark, I n o v g . Chem., 10, 1952 (1971). (20) R. D. Feltham, W. Silverthorn, a n d G. McPherson, i b i d . , 8, 344 (1969). (21) W. P . Griffith, J . C h e m Suc., 3286 (1963). (22) F. A. Cotton, R. R. Monchamp, R. J. M. Henry, and R . C. Young, J . Inovg. N u c l . Chem., 10, 28 (1959). (23) P. G. Douglas, R. D. Feltham, and H. G. Metzger, J . Amev. Chem. Soc., 93, 84 (1971). (24) R. D. Feltham, private communication. ~ . ,(1971). (25) G. Paliani, A. Poletti, and A. Santucci, J . M c l . S ~ Y ~ L8,C 63 (26) R. B. King a n d M .B. Bisnette, J . A m w . Chem. Soc., 88, 2527 (1963). (27) J. P . Collman, N. W. Hoffman, and D. E. Morris, i b i d . , 91, 5659 (1969). (28) J. P. Collman, P . Farnham, a n d G. Dolcetti, J . Amev. C h e m . Soc., 93, 1788 (1971). (29) T. Bianco, M. Rossi, and L. Uva, Inovg. Chim. A c t a , 8 , 443 (1969). (30) W, Silverthorn and R. 3.Feltham, I n o v g . Chern., 6, 1662 (1967). (31) E. P. B e r t h , S.Mizushima, T. J. Lane, and J. V. Quagliano, J . Amer. Chem. Soc., 81, 8821 (1959). (32) R . D. Feltham and R. S. h-yholm, I n o u g . Chem., 4, 1334 (1965). (33) J. P . Collman, private communication. (34) iV1. Anyoletta and G. Caglio, Gaze. Chim. Itat., 93, 1584 (1963). (35) W. Beck and K. Lottes, Bev., 98, 2657 (1965). (36) W. Hieher and K. Heinicke, 2. A n o v g . Allg. Chenz., 316, 321 (1962). Chim. I t d . , 93, 1591 (1963). (37) M. Angoletta, CQZZ. (38) J. Norton, D. Valentine, Jr., and J. P. COllmdn, J . Ameu. Chcm. Soc., 91, 7537 (1969). (39) R . D . Feltham, I n o v g . Chem., 3, 116 (1964). (40) 0. Bostrup, I n o v g . Syn., 8, 191 (1966). (41) J. M. Hollander, M. D. Holtz, T. Novakov, a n d R . L. Graham, Auk. F y s . , 28, 375 (1965); T. Yamazaki and J. M. Hollander, Nucl. Phys., 84, 505 (1966).

NOTES ordinate and formally have five or six d-electrons. I n compounds 1 4 , the metal atoms are in “abnormally low” oxidation states (0,1+, and 1+ for MO, Cr, and Mn, respectively), whereas in 5-7, the metal atoms are in “normal” oxidation states (2+ for Ru and Fe). As far as is known, the NO groups in these seven compounds are essentially linearly coordinated. The nitrogen 1s binding energies are all within A0 6 eV of 400.1 eV except for 6 and 7, for which the binding energies are 402.942and 403.3 eV, respectively. The high values for the latter compounds are indicative of relatively high positive charges on the NO groups and may be rationalized by the fact that in these compounds the metal atoms are in a “normal” oxidation state. It is noteworthy that the N-0 stretching frequencies for 6 and 7 are very high-a further indication of minimal electron donation into the x orbitals of the NO+ groups. I t is difficult to explain the low binding energy observed for 5, which contains ruthenium in a “normal” oxidation state and which has a very high N-0 stretching frequency. Perhaps the recorded binding energy corresponds to a decomposition product. 42 The second set of compounds in Table I formally have seven or eight d electrons and are either five- or s i x - ~ o o r d i n a t e . ~Most ~ of the binding energies lie within k 0 . 4 eV of 400.3 eV; the two exceptions are 10 and 15, with binding energies of 401.5 and 402.3 eV, respectively. It may be significant that both of the latter compounds are five-coordinate ; one would expect a lower electron density on the metal and the attached nitrosyl group in a five-coordinate complex than in a six-coordinate complex. However, the low binding energies of the other two five-coordinate complexes, 9 and 11, are then difficult to rationalize. I n this set of eight compounds, 15 has both the highest binding energy and the highest N-0 stretching frequency. It is interesting that, although 8 contains both terminal and bridging NO groups,26only one nitrogen 1s peak was observed. Apparently, the NO groups are so similar in electron density that they are indistinguishable by X-ray photoelectron spectroscopy. A binding energy of 402.0 eV was previously reported44 for 13, [Co(NO) (NH3)6]C12. We repeated this measurement several times with samples that were shown to be pure by magnetic susceptibility measurements and consistently observed one slightly broadened peak due to the NH3 groups (at 400.2 eV) and the NO group. By computer fitting, we were unable t o ascertain the position of the NO peak more precisely than 400.7 f 1.3 eV. It should be noted that, on going from 14 to 15, the complex ion changes only by the removal of a chloride ion-with the consequent change in coordination number from 6 to 5 . The increase in binding energy of 1.8 eV and the increase in the N-0 stretching frequency of 300 cm-1 are the expected consequences of reduced electron donation into the NO+ x orbitals. (42) Compound 6 underwent decomposition in the X ray beam, as evidenced by a gradual decrease in the 402 9-eV peak and the simultaneous growth of a 400 0-eV peak Although no similar evidence for decomposition was noticed in the case of compound 6, i t is possible t h a t such decomposition occurred before adequate counts f o r a decent spectrum were obtained (43) Compound 8 may be considered six-coordinate if the v-cyclopentadienyl ion is considered t o be tridentate. (44) D Hendrickson, J M Hollander, and W L Jolly, Inorg. Chem , 8, 2642 (1969)

Inorganic Chemistry, VoZ. 11, No. 4, 1972 895

NOTES

TABLE I NITROGEN Is BINDINGENERGIES.N-0 STRETCHING FREQUENCIES, AND M-N-0 BONDANGLESFOR SOME TRANSITION METALCOMPLEXES Compd no.

Compd

No. of d electrons"

Nitrogen 1s binding energy, eV

Nitrosyl str freq, cm-1

Re€

6 5 5 6 6 6 6

399.6 400.7 400. 7b 399.7 400.0 402. ge 403.3

1760, 1670 1690 1645 1725d 1883 1865 1939

20 20 21 22 23 24 25

11 12 13 14 15

7 8 8 8 7 8 8 8

400.7 400.3 401.5 400.7 400.0 400.7 400.5 402.3

1672, 1505 16281 1630 1676 1620 1620 1562, 1548 1852

26 27 28 29 30

16 17 18 19

8 8 7 8

401.7, 399.6 401.5, 399.7 401.2, 399.6 402.6, 400.2

1735, 1630 1720, 1650 1760 1845, 1687

1 2 3 4 5 6

7 8 9

10

M-N-0 angle, deg

176 0" 174 3d

Ref

8 9

178 3

10

120g 123h

11 12

32 32

119 0 124 4% 1743

14

28 28 30 16

179 5, 136 0

16

31

13 15

401.1 1730, 1720 33 10 400.2 17 34 10 1760-1715 163 5 400.0 35 1738 10 400.8 35 1714 10 1650 400.8 36 10 400.3 37 1615 10 180 18 1645 399.6 37 10 400. 4k 1524, 1508 10 38 399.8 17352 10 39 152 7 399.6 10 19 D Hendrickson, J. M. Hollander, and W. L Jolly, Inorg. Chem , 8, Based on the arbitrary assumption 6f N O + nitrosyl groups. Both the stretching frequency and bond angle are for the cor2642 (1969). c The bond angle refers to the corresponding C0(en)3~ salt. responding potassium salt. e Reference 42. f The stretching frequency is for the compound containing the triphenylphosphine ligand. 0 The bond angle refers to the analogous iridium complex in which one CH3 group has been replaced by an I atom. The bond angle The bond angle refers to the analogous compound in which ethylenediamine groups have refers to the analogous iridium complex. replaced the diarsine groups. 3 The bond angle refers to the analogous ruthenium complex in which two diphosphine groups have replaced the diarsine groups. b Very broad signal because of decomposition. i The stretching frequency is for the analogous bromo and iodo compounds. 20 21 22 23 24 25 26 27 28 29

Q

The third set of compounds in Table I formally have seven or eight d electrons and show two separate nitrogen Is peaks. For each of the cobalt compounds 16 and 17, two different coordination geometries are thought to be present :28 a trigonal-bipyramidal structure with a n equatorial linear nitrosyl and a squarepyramidal structure with an apical bent nitrosyl. The two nitrogen 1s binding energies observed for both of these compounds are separated by about 2 eV. Presumably the bent nitrosyl corresponds to the lower binding energy and the linear nitrosyl corresponds to the higher binding energy in each case. Two nitrogen Is binding energies were observed for 18, [Fe(NO)(diars)~] (C10&, although only one N-0 stretching frequency has been reported. Three possible explanations come to mind. Perhaps different structural forms are present (as postulated for 16 and 17), and the N-0 stretching frequency of one form is of low intensity. Perhaps the spectral splitting is due t o the paramagnetism of the sample. Finally, i t is not unlikely that the sample underwent decomposition in the X-ray beam and that the two peaks are due to the compound and its decomposition product.45 In the ruthenium complex, compound 19, two different types of nitrosyl group are bonded t o the same metal atom. l6 This square-pyramidal compound has (45) However, we were unable t o observe any evidence of a shift in the relative intensities of the peaks with time of irradiation.

an apical bent nitrosyl and a linear equatorial nitrosyl. Presumably, the higher binding energy (402.6 eV) and higher N-0 frequency (1845 cm-l) correspond t o the linear nitrosyl, and the lower binding energy (400.2 eV) and lower N-0 frequency (1687 cni-l) correspond to the bent nitrosyl. The fourth set of compounds in Table I (20-29) formally have ten d electrons and, except for 27, are fourcoordinate, Probably because of the high electron densities on the metal atoms, the binding energies are low. The low N-0 stretching frequencies are consistent with these results. We draw the overall conclusion that there is a definite correlation between the nitrogen Is binding energy of a nitrosyl group and the electron density on that group. Bent nitrosyls have low binding energies. Linear nitrosyls can have either low or high binding energies, depending on the extent of ?r back-bonding from the metal atom. The binding energies are a t least roughly correlated with the N-0 stretching frequencies : when one is high or low, so is the other. Acknowledgments.-We

are grateful t o Professors

R.D. Feltham and J. P. Collman for providing many of the samples and for helpful discussions. We thank Mr. Mark Iannone for preparing the [Co(NO)(NH3)5]Clz. This work was supported by the U. S. Atomic Energy Commission.